Subject:
|
Re: Gear spacings.
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Wed, 8 Nov 2000 01:53:48 GMT
|
Original-From:
|
Steve Baker <SJBAKER1@AIRMAIL.NETsaynotospam>
|
Reply-To:
|
sjbaker1@airmail.net&spamcake&
|
Viewed:
|
763 times
|
| |
|
|
Jennifer Clark wrote:
>
> J G Gregory wrote:
>
> > So who is going to publish the super handbook on all these? We really need
> > a "Technic Techniques" companion to the Mindstorms books. Although things
> > like this are fairly trivial to compute, having a handy reference would be
> > nice.
>
> Something else I've thought would be handy would be a program which would take
> as inputs the desired positions of the input and output axles in a gear train
> and the desired gear ratio. Space constraints in the form of "bricks" blocking
> off certain volumes of space would also be input. The program would then go off
> and compute all possible gear trains (if any) that fit the parameters. A degree
> of "fuzziness" could be built in that would allow, for example, variations of
> the desired gear ratio, and perhaps a factor could be specified stating the
> desired "perfectness" of the gear geometry.
Well, I think this is something you could divide into two parts:
STEP 1: Find all combination of gears that'll produce a gear ratio
within some user-specified range using less than some specified
number of gear wheels.
That's a simple problem to program - and it shouldn't produce *VAST* lists
of possible solutions so long as you don't let it use more than some reasonable
number of gears...there just aren't all that many combinations.
STEP 2: Find all possible ways of getting a particular set of gears to connect
and axle at point A to another axle at point B. This is also a simple
problem since there are rarely more than a couple of choices for ways
to connect one gear to it's neighbour...hence it shouldn't be hard to
simply list all possible geometries for a particular gear train - and
then reject all those that don't come close to connecting point A to
point B.
Feeding the output of STEP 1 into STEP 2 should result in a program that would
run to completion in a reasonable amount of time.
Since the number of ways to connect A to B with a gear ratio within some specified
range will typically be quite small - I think it's up to the human brain to reject
those that don't fit around the obstacles.
Since STEP 1 and STEP 2 are both useful programs in their own rights, they should
probably be left as two separate programs.
> The problem is obviously of exponential complexity, but for small numbers of
> gears is no doubt managable without resorting to complex AI or constraint
> satisfaction techniques.
I don't think you need much more than brute force techniques.
> Of course, if it could also work out the necessary bracing to hold it all
> together, it would be even better :-)
Now *that* is hard!
--
Steve Baker HomeEmail: <sjbaker1@airmail.net>
WorkEmail: <sjbaker@link.com>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sourceforge.net
http://tuxaqfh.sourceforge.net
http://tuxkart.sourceforge.net
http://prettypoly.sourceforge.net
|
|
Message is in Reply To:
 | | Re: Gear spacings.
|
| (...) Something else I've thought would be handy would be a program which would take as inputs the desired positions of the input and output axles in a gear train and the desired gear ratio. Space constraints in the form of "bricks" blocking off (...) (24 years ago, 7-Nov-00, to lugnet.robotics)
|
8 Messages in This Thread:
 
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
Active threads in Robotics
|
|
|
|
|
|
